Gas dynamics and accretion in the Galactic centre

Transcription

1 Gas dynamics and accretion in the Galactic centre Jorge Cuadra P. Universidad Católica (PUC) Chile JC acknowledges support from CONICYT- Chile through FONDECYT ( ), Basal (PFB0609) and Anillo (ACT1101) grants.

2 (From the GC group in Köln)

3 Mass losing stars in the GC Young massive stars. Many of them are Wolf-Rayet s. Mass loss rate up to ~ 10-4 M / yr each. Measured from individual spectra and confirmed by bow shocks. Gemini Observatory/Francois Rigaut

4 Data available: stellar spectra and dynamics Lu et al 2009 Eisenhauer et al 2003 Genzel et al 2003

5 Simulating the Galactic Centre Winds Use 30 Wolf-Rayet stars as gas sources. (Paumard et al 2006) Measured 2d positions and 3d velocities. Different assumptions for the 3d positions. Stellar wind properties measured for many of the stars. (Paumard et al 2001, Martins et al 2007) Typical mass loss rates ~ 10-5 M Sun / yr. Typical wind velocities ~ km/s (cooling can be important). Model using SPH code Gadget-2 (Springel 2005), starting ~1200 yr ago and let it evolve until the present time. (Cuadra, Nayakshin, et al 2005, 2006, 2008, 2015)

6 Simulated region R~1pc Bondi Radius ~0.04pc 6

7 Bondi Radius 7

8 Cooling instabilities drive clump formation. (e.g., Vishniac 1994, Burkert & Lin 2000, Calderón et al 2015) Many clumps form; a few in radial orbits produce a variable accretion rate. Possible origin for G2. (Gillessen+ 12, 13) 8 (Cuadra, Nayakshin, et al 2005, 2006, 2008, 2015)

9 Clump formation in wind collision shocks Happens due to nonlinear thin-shell instability. (Vishniac 94) Calderón+, in prep, poster Analytical models show is not as effective as it appears from SPH simulations. Still, close binaries are promising. (Calderón, Ballone, Cuadra+ 15) Currently working on AMR models.

10 Get better estimates for clump masses, orbits, and rates Calderón, PhD thesis in prep

11 Accretion rate as a function of time, measured at Rsink ~< 0.1 RBondi. Consistent with Bondi estimate from x-rays. (e.g., Baganoff+ 03)

12 stars RBondi However, temperature profiles from simulations and observations (Wang+ 13) don t match very well

13 Adding an Outflow from Sgr A* From simulations and x-ray observations, we know accretion at Bondi radius ~ Msun/yr. (e.g., Baganoff et al 2003) Faraday rotation measurements imply accretion close to horizon ~ Msun/yr. (e.g., Marrone et al 2007) => Majority of captured gas must outflow! RIAF have hot, unbound particles => convection, outflows. (e.g., Blandford & Begelman 99; Narayan+ 2000; Quataert & Gruzinov 2000) Recent X-ray observations also imply outflow. (Wang et al 2013)

14 Instantaneous mini-feedback mode Control run, no outflow Capture rate as a function of time.

15 Instantaneous mini-feedback mode Expel captured mass as soon as it enters RSch. Measure suppression of capture rate at that radius. outflow vel = 2,000 km/s Capture rate as a function of time.

16 Instantaneous mini-feedback mode Expel captured mass as soon as it enters RSch. Measure suppression of capture rate at that radius. outflow vel = 5,000 km/s Capture rate as a function of time.

17 Instantaneous mini-feedback mode Expel captured mass as soon as it enters RSch. Measure suppression of capture rate at that radius. outflow vel = 10,000 km/s Capture rate as a function of time.

18 Control run, no outflow slow outflow, km/s fast outflow, km/s None seems to fit observational constraints. (Wang+ 13) In any case, same physical conditions at Bondi radius give different capture rates there!

19 Outburst mode Sgr A* was more active in the past, as implied by X-ray echoes (e.g., Ponti+2010). Assume there was a relatively strong outflow, with mechanical power ~ erg/s. Have that outflow active for 300 yr and then turn it off.

20 Outburst mode

21 Outburst mode vel = 10,000 km/s dm/dt = 10-4 Msun/yr After ~200 yr density profile still too flat.

22 Outburst mode vel = 5,000 km/s dm/dt = 10-4 Msun/yr This model seems to match better the constraints. Notice Sgr A* would then still be recovering from past outflow.

23 X-ray Image: Models vs. Data (Russell, Wang, Cuadra 2016) pre-psf with PSF data with PSF pre-psf 4-9 kev ACIS-S/HETG 0 th order increasing outflow strength

24 X-ray Spectra: Models vs. Data (Russell+16) 2-5 ring excluding IRS 13E & PWN ACIS-S/HETG 0 th order n H =1.1e23 cm -2

25 Summary The Galactic centre provides a unique opportunity to model and observe the material feeding a SMBH. Models show that accretion rate is low, but variable at different time-scales. Formation and accretion of clumps. Related to G2? Outflows expected from SgrA* have impact on accretion and gas dynamics at Bondi radius and beyond. Observed. X-ray emission consistent with strong outflow (M~10-4 Msun/yr, v~5000km/s erg/s) ~200 yr ago, but not later.